ISSN 1980-8623

Psico http://dx.doi.org/10.15448/1980-8623.2018.4.27943

Porto Alegre, 2018; 49(4), 365-374 A rtigo O riginal

Working memory and intelligence quotient:

Which best predicts on school achievement?
Gustavo Marcelino Siquara1
Cassio dos Santos Lima2
Neander Abreu2
Escola Bahiana de Medicina e Saúde Pública, BA, Brasil
1 
2 Universidade Federal da Bahia, BA, Brasil

Abstract
Working memory (WM) refers to ability to store and manipulate information for a period of time. There is evidence of a close
relationship between WM and learning. The aim of this study was to investigate the relationship between WM, intelligence
quotient (IQ) on school achievement. Individual neuropsychological assessment was conducted following a specific protocol. The
study included 227 children aged 7 to 12 years (M = 9.87, SD = 1.34), of which 119 were females, from private and public schools
in Salvador, Bahia, Brazil. The data analysis included descriptive and inferential statistics. Academic achievement was evaluated
using the Academic Performance Test (TDE), with subtests in reading, writing and arithmetic. The components of WM were
evaluated using backward and forward Digit Span tasks and the backward and forward Corsi block-tapping task. The estimated
IQ was assessed using tasks vocabularies and block design of the Wechsler scales. The statistical tests used were linear regression
analysis (stepwise) and Pearson correlation. The results showed that the best predictors of academic performance were Digit Span
Forward, Corsi Block Backward and Digit Span Backward task. The hypothesis of non-multicollinearity was tested, and it was
verified that the constructs were independent (VIF < 10 and Tolerance > 0.20). The variance in academic performance explained
by WM model was ΔR = 0.36. High and significant correlations were observed between the components of WM and the TDE. The
WM score (Digit Span Backward + Corsi Block Backward) and the TDE were correlated (r = 0.54**). These results indicate that
WM is a good predictor of academic achievement than IQ and are consistent with other findings showing WM as a predictor of
learning or the potential for learning. This result has important implications for education, particularly with respect to intervention.
Keyword: Working memory; Academic achievement; Executive function; Gc; Intelligence quotient (IQ).

Memória de trabalho e quociente de inteligência: o que melhor prevê o desempenho escolar?

Resumo
Memória de trabalho (MT) refere-se a capacidade de armazenar e manipular informações por um período de tempo. Há evidências
de uma estreita relação entre MT e aprendizado. O objetivo deste estudo foi investigar a relação entre MT, quociente de inteli-
gência (QI) no desempenho acadêmico. A avaliação neuropsicológica individual foi realizada seguindo um protocolo específico. O
estudo incluiu 227 crianças com idades entre 7 a 12 anos (M = 9.87, SD = 1.34), das quais 119 eram do sexo feminino, de escolas
privadas e públicas em Salvador, Bahia, Brasil. A análise de dados incluiu estatística descritiva e inferencial. O desempenho
acadêmico foi avaliado usando o Teste de Desempenho Escolar (TDE), com subtestes de leitura, escrita e aritmética. Os componentes
da MT foram avaliados usando tarefas de Span de Digitos e direta e inversa e a tarefa de Blocos de Corsi versão direta e inversa.
O IQ estimado foi avaliado usando as tarefas de vocabulário e cubos das escalas Wechsler. Os testes estatísticos utilizados foram
análise de regressão linear (enter) e correlação de Pearson. Os resultados mostraram que os melhores preditores de desempenho
escolar foram as tarefas Digitos span direto, invertido e Cubos de Corsi direto e inverso. A hipótese de não-multicolinearidade foi
testada, e verificou-se que os construtos foram independentes (VIF <10 e tolerância> 0,20). A variância no desempenho acadêmico
explicado pelo modelo WM foi ΔR = 0,12. Correlações significativas foram observadas entre os componentes da MT e do TDE.
O escore da MT (Dígito span do inverso + Corsi Block inverso) e o TDE foram correlacionados (r = 0,28 **). Estes resultados
indicam que a MT é um bom preditor de realização escolar do que o QI e são consistentes com outros achados mostrando MT
como um preditor de aprendizagem ou o potencial de aprendizagem. Este resultado tem implicações importantes para a educação,
particularmente no que diz respeito à intervenção.
Palavras-chave: Memória de trabalho; Desempenho escolar; Função executiva; Gc; Quociente de inteligência (QI).

Memoria de trabajo y cociente de inteligencia: ¿cuál predice mejor el logro escolar?

Resumen
La memoria de trabajo (WM) se refiere a la capacidad de almacenar y manipular información durante un período de tiempo, y
existe evidencia de una relación cercana entre la memoria de trabajo y el aprendizaje. El objetivo de este estudio fue investigar la
relación entre la memoria de trabajo, la inteligencia (IQ) y el rendimiento escolar. La evaluación individual se realizó siguiendo un
protocolo específico. El logro escolar se evaluó mediante la prueba de Logro escolar (TDE), con subpruebas en lectura, escritura
y aritmética. Los componentes de la memoria de trabajo se evaluaron utilizando versiones anteriores y posteriores de Digit Span
y las pruebas de bloqueo de tapado de Corsi. El IQ estimado se evaluó mediante las subpruebas Vocabulary and Block Design de
Este artigo está licenciado sob forma de uma licença Creative Commons Atribuição 4.0 Internacional,
que permite uso irrestrito, distribuição e reprodução em qualquer meio, desde que a publicação
original seja corretamente citada. http://creativecommons.org/licenses/by/4.0/deed.pt_BR
Siquara, G. M., Lima, C. S., Abreu, N. | Working memory and intelligence quotient 366

Wechsler Intelligence Scale for Children III. El estudio incluyó a 227 niños de 7 a 12 años (M = 9.87, SD = 1.34), de los cuales 119
eran mujeres, de escuelas privadas y públicas en Salvador, Bahía, Brasil. El análisis de los datos incluyó estadística descriptiva
e inferencial. Las pruebas estadísticas utilizadas fueron el análisis de regresión lineal (enter) y la correlación de Pearson. Los
resultados mostraron que los mejores predictores de Logros Escolares fueron el Digit Span hacia adelante, hacia atrás y Corsi
bloquean las tareas hacia adelante y hacia atrás. Se probó la hipótesis de no multicolinealidad y se verificó que las construcciones
eran independientes (VIF < 10 y Tolerancia > 0.20). La variación en el rendimiento académico explicada por el modelo de memoria
de trabajo fue ΔR = 0.12. Se observaron correlaciones significativas entre los componentes de la memoria de trabajo y el TDE. La
puntuación de la memoria de trabajo (Digit Span Backward + Corsi Block Backward) y el TDE se correlacionaron (r = 0.28 **).
Estos resultados indican que la memoria de trabajo es un buen predictor del rendimiento escolar que el coeficiente intelectual y es
consistente con otros hallazgos que muestran que la memoria de trabajo es un predictor del aprendizaje o el potencial de aprendizaje.
Este resultado tiene implicaciones importantes para la educación, particularmente con respecto a la intervención.
Palabras clave: Memoria de trabajo; Rendimiento escolar; Función ejecutiva; Gc; Cociente de inteligencia (IQ).

Introduction either reading or math, or, most commonly, both areas

Nearly every aspect of human life depends on
memory. Individuals who cannot encode, store, or
retrieve information must rely on others for their
survival. Because learning depends on memory,
deficiencies in any aspect of memory can prevent
children and adolescents from acquiring the skills and
knowledge necessary for success in life (Dehn, 2008).
In the study of human cognitive function over the
past thirty-five years, working memory (WM) has
been one of the most influential constructs (Cowan,
2005). WM refers to the retention of information over
a brief period of time, a function that is of central
importance for a wide range of cognitive tasks and for
academic achievement (Cowan, 2005). The capacity
of WM is limited, and the imposition of excess storage
or processing demands in the course of an ongoing
cognitive activity will lead to catastrophic loss of
information from this temporary memory system
(Gathercole, Lamont, & Alloway, 2006). Children who
have difficulty manipulating mental information will
most likely have difficulty learning new subjects.
Over the past twenty years, numerous studies have
tested the hypothesis that WM is a process that forms
the basis of the ability to learn (Zheng, Swanson, &
Marcoulides, 2011; Alloway, & Passolunghi, 2011;
Alloway & Alloway, 2010; Rohde & Thompson,
2007; Gathercole, Pickering, Knight, & Stegman,
2004). WM is required whenever something is
learned because learning requires the manipulation
of information, interaction with long-term memory
and the simultaneous storing and processing of
information (Dehn, 2008). Poor WM has measurable
impacts on children’s academic performance (Holmes
& Gathercole, 2013). It is a common feature associated
with educational underachievement (Gathercole et
al., 2004), and a substantial majority of children with
poor WM skills fails to meet expected standards in
(Gathercole & Alloway, 2008). Poor WM therefore
appears to place a child at high risk of poor scholastic
attainment (Holmes & Gathercole, 2013).
Some studies have tested that WM ability, in
comparison to fluid intelligence (Gf), contributes
to better performance in reading and mathematics
(Alloway & Passolunghi, 2011; Alloway & Alloway,
2010; Alloway, 2009; Rohde & Thompson, 2007).
Alloway & Alloway (2010) showed that children’s
WM skills at 5 years of age were the best predictor of
reading, spelling, and math outcomes six years later.
Gf, in contrast, accounted for a smaller portion of the
unique variance in reading and math skills and was not
a significant predictor of spelling performance. These
results demonstrate that WM is not same construct that
Gf but rather represents a distinct cognitive skill with
unique links to learning outcomes (Alloway & Alloway
2010). Few studies have investigated the relationship
of IQ with academic achievement. This suggests that
WM deficiencies are associated with poor learning
outcomes and are a high risk factor for school failure
in children (Alloway, Gathercole, & Elliott, 2010).
The majority of the studies comparing the predictive
power of WM for academic performance are used to
use the concept of Gf (Alloway & Passolunghi, 2011;
Alloway & Alloway, 2010; Alloway, 2009), compared
with studies that examine the predictive power for
academic achievement from IQ.
WM capacity is usually measured by utilizing
complex memory span tasks in which children
must simultaneously store and process information
(Gathercole & Alloway, 2004). Gathercole, Pickering,
Knight & Stegmann (2004) found that WM skills
were excellent predictors of whether children would
obtain low, average or high scores on both English
and math assessments at Key Stage 1 (6-7 years) and
math assessments at Key Stage 3 (13-14 years). The
study showed that WM, more than intelligence, is an

Psico (Porto Alegre), 2018; 49(4), 365-374

Siquara, G. M., Lima, C. S., Abreu, N. | Working memory and intelligence quotient 367

excellent predictor of children’s school achievement
across domains (Lu, Weber, Spinath, & Shi, 2011). This
is also consistent with previous studies reporting that
the specific associations between WM and achievement
remain when controlling for intelligence (Alloway,
2009; Swanson, Jerman, & Zheng, 2008; Maehler &
Schuchardt, 2009). Concerning the relationship between
WM and intelligence, the moderate inter-correlation of
.36 is in line with the view that the two constructs are not
isomorphic (Lu, Weber, Spinath, & Shi, 2011).
Academic achievement and development of
verbal skills, such as reading decoding, reading
comprehension, mathematics, and written expression,
depend heavily on the adequate functioning of WM
(Dehn, 2008). The strong relations between specific
areas of academic achievement and short-term and WM
components are well established (Berninger & Richards,
2002; Swanson, 2001). Common classroom activities
that impose simultaneous demands on storage and
processing include listening to a speaker while trying
to take notes, following complex instructions, decoding
unfamiliar words, spelling sentences from memory
and mental arithmetic. Learning is reduced, or at least
slowed, when available WM capacity is reduced through
overloading or dividing attention (Dehn, 2008).
Evaluation of WM development in the early school
years and intervention programs for children with
learning difficulties may reduce academic impacts.
The research should investigate the cognitive functions
that are key to academic performance to understand
better the variables that influence the acquisition of
new knowledge.
The aim of this study was to investigate the
relationship between WM, IQ and academic
achievement. In this study, we investigated whether
WM or IQ is the best predictor of student academic
performance as well as the relationship between WM
with IQ. The hypothesis tested in the study is that
more highly developed WM is associated with better
academic achievement.
Methods
Participants
The study included 227 children between 7 and 12
years old (mean = 9.87, SD = 1.34), with 119 females
and 108 males. All children were private or public
school students between the 2nd and 5th grades in
Salvador, Bahia, Brazil.

Table 1
Socioeconomic characteristics of the study participants

Educational Level Educational Level Reading Habits

Attended Preschool Monthly Income
of Mother of Father of Children
(%) (%)
(%) (%) (%)
No 5.1
1 to 2 years 14.7
3 to 4 years 80.3
Incomplete Elementary School 3.6
Elementary School I 1.8 2.9
Elementary School II 10.9 4.8
School 22.2 8.7
Graduation 24.9 31.4
Specialization / Masters 35.3 25.1
Doctorate 1.4 1.4
R$ 151-300 1.5
R$ 301-600 5.3
R$ 600-1.200 17.7
R$ 1.200-2.400 7.2
R$ 2.400-4.800 7.2
R$ 4.800-6.000 6.7
R$ 6.000-10.000 23.0
R$ Above 10.000 31.6
Always 22.5
Sometimes 24.8
Frequently 47.2
Never 5.5

Psico (Porto Alegre), 2018; 49(4), 365-374

Siquara, G. M., Lima, C. S., Abreu, N. | Working memory and intelligence quotient 368
Table 1 displays the socioeconomic profile of
participants. Most studied in preschool for at least 3
years (80.3%). Parents generally had a middle level of
education. The monthly income of the participants was
also good distribution.
Procedures
The parents consented to the participation of their
children in the study, i.e., they signed a consent form
and received a copy of the document. Children who
had been authorized underwent neuropsychological
assessment. Individual neuropsychological assessment
was conducted following a specific protocol. All testing
occurred during school hours and was conducted
in the classes of the participants, with an estimated
duration of 1 hour and a half. The parents were sent
a socioeconomic questionnaire. This questionnaire
aimed to assess whether the child took medication or
had any medical problems, such as prematurity, low
birth weight or psychiatric or neurological illness. To
participate in the study, children could not present any
health problems.
Data Analysis
The data analysis used descriptive and inferential
statistics. The descriptive statistics included the mean,
standard deviation, and range. The inferential statistics
included bivariate correlations and linear regression
analysis. The software used was SPSS version 20
(IBM, 2011).
Instruments
Intelligence (IQ Estimated). The IQ of the children
was estimated using the vocabulary and cube tasks
of the Wechsler Intelligence Scale for Children, 3rd
Version (WISC-III) (Wechsler, 2002). Mello et al.
(2011) reported on the feasibility of estimating a child’s
IQ through these two tasks.
Vocabulary (WISC-III). The child is asked to
provide settings for a list of 35 words that are read one
by one.
Block Design (WISC-III). The child is given 16
white and red blocks and is shown figures made up of
white and red cubes. The child is asked to assemble
the blocks to correspond to the figure, a task requiring
mental decomposition and rotation of the figure.
Short Term Memory (STM) and Working Memory
(WM). To evaluate STM and WM, the digit span task
from the WISC-III and the Corsi Block-tapping Test
were used, both forward and backward versions of the
tasks.
Digit Span (WISC-III). The examiner reads
aloud a sequence of numbers. For the forward version
Psico (Porto Alegre), 2018; 49(4), 365-374
of the task, the child repeats the numbers in the
same order in which they were spoken (assessment
of STM). For the backward version, the child
repeats the numbers in reverse order (assessment of
WM).
Corsi Block Test. The task includes nine cubes
mounted on a board (Lezak, 1995). The examiner
touches the blocks in a specific sequence and asks the
child to repeat the sequence in the same order for the
forward version of the task (assessment of STM). For
the backward version, the examiner touches the blocks,
forming a sequence, but the child must touch them in
reverse order (assessment of WM).
Academic achievement
Academic achievement was assessed with a
Brazilian instrument, the “Teste de Desempenho
Escolar” (TDE). The TDE is a psychometric instrument
with individual application to the evaluation of the
fundamental capabilities of school performance,
specifically spelling, arithmetic and reading (Stein,
1994). It is designed for evaluation of students from
1st to 6th grade of elementary school, although it can
be used with some reservations for the 7th and 8th
grades. It has been validated in Brazil using a sample
from Porto Alegre. This test evaluates the fundamental
skills of students using the following tasks: 1. Spelling:
Students listen to and write 34 single words presented
in dictation form; 2. Arithmetic: Students are given 3
mathematical problems in oral form and 35 written
arithmetic calculations, fora total of 38; 3. Reading:
Students are asked to recognize 70 isolated words in
context. The total score for each task is obtained from
the sum of each correct answer that the participant has
obtained (Stein, 1994).
Results
Table 2 shows the performance of the children on
the WM tasks and the TDE. The data in Table 2 present
the skewness and kurtosis as indicators of the normal
tendency of the tasks in the sample. In all tasks, the
skewness was smaller than one and the kurtosis closer
to zero.
Table 3 shows the children’s performance for each
task, comparing age groups. It can be observed that
STM and WM improve with age, and the performances
of the children on the TDE also improve with age.
Despite this improvement in performance, it was not
possible to separate the participants into homogenous
groups using a post-hoc ANOVA. The mean estimated
IQ was above average (110), indicating a medium
intelligence level for the participants.
Siquara, G. M., Lima, C. S., Abreu, N. | Working memory and intelligence quotient 369

Table 2
Descriptive statistics of row scores for measures cognitive

Min. Max. Skewness Kurtosis Mean SD

Digit Span Forward 3 15 1.063 1.856 7.51 2.04
Digit Span Backward 1 11 .817 1.226 4.49 1.62
Corsi Block Test Forward 3 12 -.188 .553 7.68 1.53
Corsi Block Test Backward 1 12 -.359 -.359 7.04 2.18
TDE Total 3.00 138.00 -1.835 -3.699 105.34 27.9
WM Total 2 20 -.026 -.029 12.15 3.01
WM Total (Digit Span Backward + Corsi Block Test Backward).
TDE Total (Arithmetic + Reading + Spelling).

Table 3
Performance for age in each of the cognitive tasks

Component Short-Term Memory Working Memory (WM)

Visuo-Spatial
Sub-component Verbal Visuo-Spatial Loop Phonological
Sketchpad WM Total TDE Total
Digit Span Corsi Block Test Digit Span Corsi Block Test
Age N
Forward Forward Backward Backward
7 7 7.14(0.9) 6.71(0.7) 3.86(1.0) 5.29(2.4) 9.14(2.5) 87.00(7.3)
8 18 7.06(0.9) 7.22(1.6) 4.22(1.7) 5.22(1.7) 9.44(2.7) 101.94(8.6)
9 37 7.54(1.5) 7.86(1.1) 4.32(1.3) 7.16(2.2) 11.48(2.7) 110.51(9.45)
10 45 8.02(1.6) 7.64(1.1) 4.69(1.4) 7.76(1.6) 12.44(2.4) 118.40(8.87)
11 27 7.4(1.4) 8.44(1.7) 5.39(1.6) 8,19(1.7) 13.62(2.6) 124.70(13.92)
12 14 9.57(2.0) 9.00(2.0) 5.57(1.4) 9,43(1.2) 15(1.9) 132.00(5.39)
Total 148 7.79(1.6) 7.88(1.4) 4.72(1.5) 7.42(2.1) 12.14(3.0) 115.46(14.29)
WM Total (Digit Span Backward + Corsi Block Test Backward).
TDE Total (Arithmetic + Reading + Spelling).

Table 4 shows the correlations between the tasks of
the TDE and the measures of memory. All correlations
between the reading, spelling and mathematics tasks
and the memory measures were significant. The
Arithmetic Task had the highest correlation with
performance on the Corsi Block Backward (r =.59,
p ≤ 0.01). The Reading Task showed the strongest
correlation with the Digit Span Forward task (r =.43,
p ≤ 0.01). For the Spelling Task, the correlations were
similar among the Corsi Block Backward, Digit Span
Forward, and Digit Span Backward tasks (r =.41, r =.48
and r =.37, respectively, all with p≤0.01).

Table 4
Bivariate correlational analysis between school performance, short-term memory and working memory

1 2 3 4 5 6 7 8 9 10
1. Corsi Block Forward –
2. Corsi Block Backward .451** –
3. Digit Span Forward .268** .376** –
4. Digit Span Backward .269** .361** .443** –
5. WM Total .451** .879** .488** .762** –
6. QI Estimated .235** .351* .355** .291** .392* –
7. Arithmetic .377** .590** .450** .406** .617** .324** –
8. Reading .252** .332** .436** .376** .423** .424** .545** –
9. Spelling .299** .413** .482** .371** .475** .435** .684** .817** –-
10. TDE Total .336** .473** .506** .428** .546** .448** .785** .929** .934** –
* p ≤ 0.05; **p ≤ 0.01.
WM Total (Digit Span Backward + Corsi Block Test Backward).
TDE Total (Arithmetic + Reading + Spelling).

Psico (Porto Alegre), 2018; 49(4), 365-374

Siquara, G. M., Lima, C. S., Abreu, N. | Working memory and intelligence quotient 370

Of interest was the finding (Table 4) that academic The correlations between the estimated IQ measure
achievement (TDE Total) showed the highest and academic achievement (TDE) were significantly.
positive correlations with the WM tasks (Corsi Block All correlations was a medium power include TDE
Backward, r =.47, and Digit Span Backward, r =.42, total (r=.54, p ≤ 0.01). For the other measures of TDE
p ≤ 0.01). Another strong positive correlation occurred estimated IQ also showed good correlations.
between TDE Total and WM (r =.56, p ≤ 0.01) and the Table 5 shows the results of the linear regression
Arithmetic WM task (r =.54, p ≤ 0.01). Interestingly, model using the STM tasks (Digit Span Forward and
the task of Digit Span Forward showed higher positive Corsi Block Forward) and the WM tasks (Corsi Block
correlations with the Spelling and Reading tasks (r =.48 Backward and Digit Span Backward) to predict total
and r =.43; p ≤ 0.01); however, when we analyzed TDE (Arithmetic + Reading + Spelling). A stepwise
overall academic achievement (TDE total), the Digit method was used to build the regression model. The
Span Backward task had a higher correlation (r =.42; results of the construction of the model indicate that
p ≤ 0.01). the tasks that best explained the TDE variance were
Digit Span Forward, Corsi Block Backward and Digit
Table 5 Span Backward. The greatest variance was explained
Linear regression to predict academic achievement (TDE specifically by the WM tasks. The Digit Span Forward,
total) from WM task Corsi Block Backward and Digit Span Backward,
Std. explained together (ΔR=.36) of the TDE total. The
R2 ΔR B Error β Corsi Block Forward task did not significantly
B explain any variance in the model, so it was then
Step 1 .25 .25 excluded in the stepwise method. The hypothesis of
Constant 52.344 6.20 non-multicollinearity was tested, and it was verified
Digit Span Forward 3.366 .483 .508** that the constructs were independent (VIF<10 and
Step 2 .35 .34 Tolerance>0,20). The requirements of linearity and
Constant 35.86 6.51 homoscedasticity were met.
Digit Span Forward 5.29 .489 .382** The IQ submitted an variance explained less
Corsi Block Backward 4.19 .662 .328** than WM tasks, but showed a significant prediction
Step 3 .37 .36 (ΔR=.19). The hypothesis of non-multicollinearity
Constant 31.81 6.52
was tested, and it was verified that the constructs were
Digit Span Forward 4.44 .839 .761**
independent (VIF<10 and Tolerance>0,20).
Corsi Block Backward 3.64 .753 .806**
Table 7 shows the results of the linear regression
Digit Span Backward 3.17 1.034 .776**
model using only Corsi Block Forward to predict
* p ≤ 0.01; **p ≤ 0.001. Arithmetic TDE. The data showing above is interesting
because only one task which assesses visuospatial WM
Table 6 is predicting a good value for the test of arithmetic
Linear regression to predict academic achievement (TDE total) TDE.
from IQ estimated (Gc)

R2 ΔR B
Std. Error
β Discussion
B
Step 1 .19 .19 The aim of the study was to analyze the correlational
Constant 30.906 10.160 and predictive power of WM and estimated IQ with
IQ Estimated .671 .0.91 .444** regard to academic achievement. The findings of the
* p ≤ 0.01; **p ≤ 0.001. study indicate good correlations between WM and
academic achievement, more specifically, reading,
Table 7 spelling and mathematics. Previous published studies
Linear regression to predict performance in also indicate these correlations between WM and
sub-test arithmetic from visuospatial working memory
academic achievement (Alloway, 2009; Alloway,
Std. Error 2006; Rasmussen & Bisanz, 2005; Cowan & Alloway,
R2 ΔR B β
B 2008). Overall correlations between WM measures
Step 1 .30 .30 and achievement range as high as .55 to .92 (Swanson,
Constant 6.54 1.98 1999). There is a strong empirical foundation for the
Corsi Block Backward 2.03 .483 .554** view that learning difficulties reflect a fundamental
* p ≤ 0.01; **p ≤ 0.001. deficit in WM (Swanson & Siegel, 2001).

Psico (Porto Alegre), 2018; 49(4), 365-374

Siquara, G. M., Lima, C. S., Abreu, N. | Working memory and intelligence quotient 371
The current study found results consistent with
the literature, i.e., the tasks that had good correlations
with academic achievement were WM tasks. WM
requires executive resources for the processing
aspect of the task. The link suggests that the ability to
mentally store and manipulate information may allow
a greater capacity for learning, explaining the higher
correlations and prediction model between academic
achievement and WM tasks. Interestingly, the analysis
showed one of the largest correlations between the total
score of the WM tasks and the total score on reading,
spelling and mathematics. This may indicate that the
ability to manipulate the information actually improves
verbal and spatial learning ability. A higher correlation
between verbal aspects from WM and tasks of reading
and spelling was observed. This may be related to the
development of components of WM (phonological
loop and visuospatial sketch) with the greatest ability
to manipulate verbal and spatial information.
Another task used which showed excellent
relationships with performance in arithmetic was the
task of Corsi Block Backward. Only this task showed
a variance explained (ΔR=30). These findings may
indicate that the ability of visuospatial manipulation
can predict performance in mathematics in children.
Studies such as Meyer, et al. (2010) argue that tasks
involving the central executive of working memory
facilitate the learning of mathematics.
The Corsi Block task showed a good prediction
on arithmetic performance. It is likely that this result
is explained by a greater capacity of visuospatial
manipulation. With the further development of the
visuospatial sketch the child can accomplish more
manipulations and mental rotations from visual stimuli
as presented on mathematical tasks. Some studies in the
literature indicate the importance of the development
of the visuospatial sketch for the mathematical tasks
(Andersson& Lyxell, 2006, Zheng, Swanson, &
Marcoulides, 2011). This finding may contribute to the
identification of mathematical problems..
Another important finding was that WM improves
with increasing age. This demonstrates that storage
and manipulation of information also increases.
Other studies also report similar findings (Gathercole,
Alloway 2008; Alloway 2006), with implications
especially in the classroom. However, this pattern was
not observed in homogeneous groups using a post-hoc
ANOVA according to age. Perhaps WM develops along
a continuum with no specific landmarks in accordance
with age. These findings point to the importance of
the development of WM for academic achievement
and learning. Many studies now indicate failures in
the development of WM in children with learning
Psico (Porto Alegre), 2018; 49(4), 365-374
difficulties (Alloway, 2006; Lu et al. 2011; Alloway,
Gathercole, Elliott, 2010; Gathercole, 2008; Henry &
Winfiel, 2010; Meyer, Salimpoor, Wu, Geary, Menon,
2010).
The estimation of IQ was made through the WISC-
III tasks. These tasks assess crystallized intelligence
and are associated with specific aspects of learning.
This intelligence was developed from cultural and
educational experiences, present in most school
activities (Cronbach, 1996). The findings of the study
indicate that IQ (Gc) has significant relationship to
academic achievement. Many studies have assessed
fluid intelligence, and correlations are lower when
compared with WM (Unsworth &Engle, 2005; Lu et
al. 2011; Hornung, Brunner, Reuter, Martin, 2011).
However Gc and predictions showed good relations
with academic performance, although these relations
are weaker than the WM.
In the regression model, TDE was explained
mainly from the WM tasks. The amount of variance
in performance on the TDE that was explained by the
tasks used was 36%. Interestingly the explanation of
academic achievement was mainly due to the WM tasks.
This indicates the strength of WM tasks in predicting
overall performance on the TDE. Other studies, such
as that by Alloway (2009), included regression models
that also indicated strong relationships between WM
and academic achievement. The findings from the
regression analyses indicated that both WM capacity
and prior domain-specific knowledge were unique
predictors of learning outcomes 2 years later. The
finding that initial reading and math skills are important
precursors to subsequent learning fits well with the
existing literature (Swanson, 2006; Butterworth,
2005). IQ was not a significant predictor of learning
outcomes once WM capacity and prior knowledge
were statistically controlled. It is also of interest that
working-memory capacity predicted subsequent
skills in both reading and math (Alloway, 2009). The
evidence from the study indicates that WM is a good
predictor of academic achievement, as measured by
TDE tasks.
The findings of this study are consistent with
others indicating that poor WM is associated with poor
academic attainment. It fits in with the notion that WM
is the foundation of learning (Martinussen, Hayden,
Hogg-Johnson, Tannock, 2005) and is associated
with learning disabilities regardless of the core deficit
(Nigg, 2006). Early screening of WM problems can
prevent subsequent learning difficulties (Alloway,
Gathercole & Elliott, 2010). While there are a range
of cognitive skills that are crucial to learning, there is
growing evidence that WM is one of the best predictors
Siquara, G. M., Lima, C. S., Abreu, N. | Working memory and intelligence quotient 372

of attainment (Alloway, Banner &Smith, 2010). The
identification of this strong relationship between WM
performance and academic studies is interesting and
could lead to development of interventions to stimulate
WM. Training programs that directly target WM
provide important evidence that it is possible to make
enduring changes to these memory abilities. Cogmed
Working Memory Training provides intensive practice
on a range of computer-based memory tasks over
20–25 sessions (Klingberg, Forssberg, & Westerberg,
2002; Klingberg et al., 2005). There is also preliminary
evidence of accelerated learning following such
training, with significant improvements in math scores
reported several months after training for children
with WM impairments (Holmes, Gathercole, Dunning,
2009) and improvements in reading comprehension
reported post-training for children with special
educational needs (Dahlin, 2010).
In summary, the data from the research suggests
that WM predicts academic achievement, specifically
performance in reading, spelling and arithmetic.
The ability of visuospatial manipulation seems to
be closely linked to performance in mathematics.
The results of the regression model indicate that the
ability to manipulate information is intimately linked
to learning. Compared to WM, estimated IQ (Gc)
showed lower correlations and predictive power of
academic achievement. Finally, this study indicates
that the best the development of WM, the best the
school performance in children.

References
Alloway, T. P. (2006). How does working memory work in the classroom? Educational Research and Reviews, 1(4),
134-139.
Alloway T. P. (2009) Working memory but not IQ predicts sub- sequent learning in children with learning difficulties.
European Journal Psychology Assessment, 25, 92–98. https://doi.org/10.1027/1015-5759.25.2.92
Alloway, T. P. Banner, G. E. & Smith, P. (2010). Working memory and cognitive styles in adolescents’ attainment. British
Journal of Educational Psychology, 80, 567-581. https://doi.org/10.1348/000709910X494566
Alloway, T. P. & Passolunghi, M. C. (2011).The relationship between working memory, IQ, and mathematical skills in
children. Learning and Individual Differences, 21(1), 133-137. https://doi.org/10.1016/j.lindif.2010.09.013
Alloway, T. P., Gathercole, S. E., & Elliott, J. (2010).Examining the link between working memory behaviour and
academic attainment in children with ADHD. Developmental medicine and child neurology, 52(7), 632-6. https://doi.
org/10.1111/j.1469-8749.2009.03603.x
Alloway, T. & Alloway, R. G. (2010).Investigating the predictive roles of working memory and IQ in academic
attainment. Journal of Experimental Child Psychology, 106(1), 20-29. Elsevier Inc. https://doi.org/10.1016/j.
jecp.2009.11.003
Atkinson, R. C. & Shiffrin, R. M. (1968). Human memory: a proposed system and its control processes. In The Psychology
of Learning and Motivation: Advances in Research and Theory (pp. 89-195). New York: Academic. https://doi.
org/10.1016/S0079-7421(08)60422-3
Berninger, V. W. & Richards, T. L. (2002). Brain literacy for educators and psychologists. New York: Academic Press.
Butterworth, B. (2005). The development of arithmetical abilities. Journal of Child Psychology and Psychiatry, 46, 3-18.
https://doi.org/10.1111/j.1469-7610.2004.00374.x
Cronbach, L. J. (1996). Fundamentos da testagem psicológica. Porto Alegre: Artes Médicas.
Cowan, N. (2005). Working Memory Capacity.New York: Lawewnce Erlbaum.
Cowan, N. & Alloway, T. P. (2008). The development of working memory. In N. Cowan (Ed.), Development of memory
in childhood, Hove: Psychology Press.
Dahlin, K. (2010). Effects of working memory training on reading in children with special needs. Reading and Writing,
24, 479-491. https://doi.org/10.1007/s11145-010-9238-y
Dehn, M. J. (2008) Working memory and academic learning: assessment and intervention. Jonh Wiley & Sons, Hoboken,
New Jersey.
Gathercole, S. E. (2008). Working memory in the classroom. Psychologist, 21, 382-385.
Gathercole, S. E. & Alloway, T. P. (2004). Working memory and classroom learning. Dyslexia Review, 15, 4-10.
Gathercole, S. E. & Alloway, T. P. (2008). Working memory and learning: A guide for teachers. Sage: UK.
Gathercole, S. E., Lamont, E. & Alloway, T. P. (2006) Working memory in the classroom. In S. J. Pickering (Ed.).
Working memory and Education (pp. 219-240). Burlington, MA: Academic Press. https://doi.org/10.1016/B978-
012554465-8/50010-7
Gathercole, S. E., Pickering, S. J., Knight, C., & Stegmann, Z. (2004).Working memory skills and educational attainment:
Evidence from National Curriculum assessments at 7 and 14 years of age. Applied Cognitive Psychology, 40, 1-16.
https://doi.org/10.1002/acp.934

Psico (Porto Alegre), 2018; 49(4), 365-374

Siquara, G. M., Lima, C. S., Abreu, N. | Working memory and intelligence quotient 373

Gathercole, S. E., Pickering, S.J., Knight, C., & Stegman, Z. (2004).Working memory skills and educational attainment:
Evidence from National Curriculum assessments at 7 and 14 years of age. Applied Cognitive Psychology, 18, 1-16.
https://doi.org/10.1002/acp.934
Henry, L. Winfiel, J. (2010). Working memory and educational achievement in children with intellectual disabilities.
Journal of Intellectual Disability Research, 54, (4), 354-365. https://doi.org/10.1111/j.1365-2788.2010.01264.x
Holmes, J. & Gathercole, S. E. (2013).Taking working memory training from the laboratory into schools, Educational
Psychology. An International Journal of Experimental Educational Psychology. https://doi.org/10.1080/01443410.
2013.797338
Holmes, J., Gathercole, S., E., Dunning, D. L. (2009). Adaptive training leads to sustained enhancement of poor working
memory in children. Development Science, 12, 9-15. https://doi.org/10.1111/j.1467-7687.2009.00848.x
Hornung, C., Brunner, M., Reuter, R. A. P., & Martin, R. (2011).Children’s working memory: Its structure and relationship
to fluid intelligence. Intelligence, 39(4), 210-221. https://doi.org/10.1016/j.intell.2011.03.002
Jarrold, C. Towse, J. N. (2006). Individual Differences in Working Memory. Neuroscience, 139, 39-50. https://doi.
org/10.1016/j.neuroscience.2005.07.002
Klingberg, T., Fernell, E., Olesen, P. J., Johnson, M., Gustafsson, P., Dahlstrom, K. Westerberg, H. (2005). Computerised
training of working memory in children with ADHD-a randomised, controlled trial. Journal of the American Academy
of Child and Adolescent Psychiatry, 44, 177-186. https://doi.org/10.1097/00004583-200502000-00010
Klingberg, T., Forssberg, H., & Westerberg, H. (2002).Training of working memory in children with ADHD. Journal of
Clinical and Experimental Neuropsychology, 24, 781-791. https://doi.org/10.1076/jcen.24.6.781.8395
Lezak, M. D. (1995) Neuropsychological Assessment, Oxford University Press.
Lu, L. Weberc, H. S. Spinath, F. M., Shi, J. (2011) Predicting school achievement from cognitive and non-cognitive
variables in a Chinese sample of elementary school children. Intelligence, 39, 130-140. https://doi.org/10.1016/j.
intell.2011.02.002
Maehler, C. & Schuchardt, K. (2009). Working memory functioning in children with learning disabilities: does
intelligence make a difference? Journal of Intellectual Disability Research, 53(1), 3-10. https://doi.org/10.1111/j.1365-
2788.2008.01105.x
Martinussen R, Hayden J, Hogg-Johnson S, Tannock R. (2005) A meta-analysis of working memory impairments in
children with attention-deficit ⁄ hyperactivity disorder. Journal American Academic Child Adolescent Psychiatry, 44,
377-84. https://doi.org/10.1097/01.chi.0000153228.72591.73
Mello, C. B., Argolo, N., Shayer, B. P. M, Abreu, N., Godinho K. et al. (2011). Versão Abreviada do WISC-III: Correlação
entre QI Estimado e QI Total em Crianças Brasileiras. Psicologia: Teoria e Pesquisa, 27(2).
Meyer, M. L, Salimpoor, V.N., Wu, S.S., Geary, D. C., Menon, V. (2010).Differential contribution of specific working
memory components to mathematics achievement in 2nd and 3rd graders. Learn Individual Difference, 20: 101-109.
https://doi.org/10.1016/j.lindif.2009.08.004
Mukunda, K.V., Hall V.C. (1992) Does performance on memory for order correlate with performance on standardized
measures of ability? A meta analysis. Intelligence 16: 81–97. https://doi.org/10.1016/0160-2896(92)90026-N
Nigg, J. (2006) What Causes ADHD? Understanding What Goes Wrong and Why. New York: Guilford Press.
Rasmussen, C. & Bisanz, J. (2005).Representation and working memory in early arithmetic. Journal of experimental
child psychology, 91(2), 137-57. https://doi.org/10.1016/j.jecp.2005.01.004
Rohde, T. E. & Thompson, L. A. (2007). Predicting academic achievement with cognitive ability. Intelligence, 35(1),
83-92. https://doi.org/10.1016/j.intell.2006.05.004
Stein, L. M. (1994). Teste de Desempenho Escolar: manual para a aplicação e interpretação. São Paulo, Casa do
Psicólogo.
Swanson, H. L. (1999). What develops in working memory? A life span perspective. Developmental Psychology, 35(4),
986-1000. https://doi.org/10.1037/0012-1649.35.4.986
Swanson, H. L. (2001). Are working memory deficits in readers with learning disabilities had to change? Journal of
Learning Disabilities, 33, 551-566. https://doi.org/10.1177/002221940003300604
Swanson, H. L. & Beebe-Frankenberger, M. (2004).The relationship between working memory and mathematical
problem solving in children at risk and not at risk for math disabilities. Journal of Education Psychology, 96,
471-491. https://doi.org/10.1037/0022-0663.96.3.471
Swanson, H. L. & Jerman, O. (2006). Math disabilities: A selective meta-analysis of the literature. Review of Educational
Research, 76, 249-274. https://doi.org/10.3102/00346543076002249
Swanson, H. L. & Siegel, L. (2001). Learning disabilities as a working memory deficit. Issues in Education: Contributions
from Educational Psychology, 7, 1-48.
Swanson, H. L., Jerman, O., & Zheng, X. (2008).Growth in working memory and mathematical problem solving in
children at risk and not at risk for serious math difficulties. Journal of Educational Psychology, 100, 343-379. https://
doi.org/10.1037/0022-0663.100.2.343
Swanson, H. L. (2006). Working memory and reading disabilities: Both phonological and executive processing deficits
are important. In T. P. Alloway & S. E. Gathercole (Eds.), Working memory in neurodevelopmental conditions
(pp. 59-88). East Sussex, UK: Psychology Press.

Psico (Porto Alegre), 2018; 49(4), 365-374

Siquara, G. M., Lima, C. S., Abreu, N. | Working memory and intelligence quotient 374

Unsworth, N. & Engle, R. W. (2007). On the division of short-term and working memory: An examination of simple
and complex spans and their relation to higher-order abilities. Psychological Bulletin, 133, 1038-1066. https://doi.
org/10.1037/0033-2909.133.6.1038
Wechsler, D. (2002). Wisc-III. Escala de Inteligência para Crianças. (V. L. Figueiredo, Trad.). São Paulo: Casa do
Psicólogo. (Original publicado em 1949).
Zheng, X., Swanson, H. L., & Marcoulides, G. A. (2011). Working memory components as predictors of children’s
mathematical word problem solving. Journal of experimental child psychology, 110(4), 481-98. Elsevier Inc. https://
doi.org/10.1016/j.jecp.2011.06.001

Dados dos autores:

Gustavo Marcelino Siquara – Mestre, Escola Bahiana de Medicina e Saúde Pública.
Cassio dos Santos Lima – Mestrando, Universidade Federal da Bahia.
Neander Abreu – Doutorado, Universidade Federal da Bahia.

Endereço para correspondência:

Gustavo Siquara
Av. Adhemar de Barros, s/n, Pavihão 04, ISP – Ondina
41170-110 Salvador, BA, Brasil
<gustavosiquara@hotmail.com>

Recebido em: 10.07.2017

Aceito em: 09.04.2018

Psico (Porto Alegre), 2018; 49(4), 365-374